Difference between revisions of "VVV15 projects"

From Wiki for iCub and Friends
Jump to navigation Jump to search
Line 38: Line 38:

Link of the video of the demo :
[https://drive.google.com/file/d/0B-9Q6q_cNomFdk5FYXdvc1dpejQ/view?usp=docslist_api Video of the demo]

= Analysis and Control of iCub joint elasticity =
= Analysis and Control of iCub joint elasticity =

Latest revision as of 01:01, 31 July 2015

Add here an entry for each project, please nominate a responsible person for each project.

IOL with Android in the Loop

Responsible: Alessandro Roncone

Participants: Alessandro Roncone, Vadim Tikhanoff, others

Short description: we have recently interfaced Android with YARP. We would like to modify the interactive object learning scenario so that the robot can be controlled with your phone. IOL in the loop, details.

Object Affordances Exploration

Responsible: Francesca Stramandinoli

Participants: Francesca, Francesco N., Ugo, Vadim, Daniele D.

Short description: Build an application that allows the iCub to explore object affordances

Tactile perception

Responsible: Nawid Jamali

Participants: Nawid Jamali, Massimo Regoli and Takato Horii

Short description: use iCub's tactile sensors to extract contact-location independent features, explore and grasp an object.

Event-driven iCub development

Responsible: Arren Glover

Participants: Arren, Chiara, Marko, Samantha, Valentina

Short description: use iCub's event driven sensors to perform ball tracking, integrating modules for ego-motion compensation, SpiNNaker attention and face detection.

Proactive Tagging

Responsible: Tobias Fischer

Contributors: T. Fischer, M. Petit, A.-L. Mealier, U. Pattacini, C. Moulin-Frier, J. Puigbo, J. Copete, B. Higy

Short description: The aim is to learn language proactively. There will be different `drives` 1) exploring, 2) triggering questions for labeling unknown objects in the objects properties collector. Pro-active tagging will be triggered because of either an (at the moment) unachievable goal, or because of the drive to explore.

File:WYSIWYD VVV15.pdf

Link of the video of the demo : Video of the demo

Analysis and Control of iCub joint elasticity

Responsible: Nuno Guedelha

Participants: Nuno Guedelha, Daniele Pucci

Short description: The goal is to identify the joint flexibility parameters and integrate them in the motion controller model, for improving stability and accuracy. For the analysis, validation and performance evaluation, we will focus on the use case of a simple trajectory tracking.

File:VVV15 Task11 HandlingJointFlexibility.pdf

The crawling iCub

Responsible : Serena Ivaldi

Group members : Marie Charbonneau, Dorian Goepp

There has been code used to make iCub crawl using Central Pattern Generators. We would like to use this code, port it to the current YARP version and use a learning/optimisation loop to tune the CPG's parameters. We shall detail this more on the project's page.

Object tracking in 3D

Responsible : Vadim Tikhanoff

Group members : Lilita Kiforenko

Short description: Detect and track some objects in 3D using pcl particle filter.

Balancing on a seesaw with the Purpley

Responsible : Daniele Pucci

Group members : Francesco Romano, Silvio Traversaro, Nuno Guedelha

We would like to carry on the first experiments of balancing on a seesaw with iCubGenova02, also known as the "Purpley". For the expected results, see


Bayesian whole-body dynamics estimation

Responsible : Francesco Nori

Group members : Claudia Latella, Silvio Traversaro, Francesco Nori

Group slides : File:VVV15-berdy.pdf

We would like to replicate the wholeBodyDynamicsTree module (for internal and external torque estimation module) within BERDY, which performs the same estimation in presence of multiple redundant measurements. Estimation is performed as a maximum-a-posteriori (MAP) strategy and it is framed in a Bayesian framework.

Analyzing the motion of iCub

Responsible : Alessia Vignolo

Group members : Roberto Barone, Carlos Cardoso, Alessia Vignolo

We would like to design some robot movements (artificial and natural ones) and analyze them using the informations taken from the encoders and the cameras.